A maintenance policy is proposed for a degrading system, and a gamma process is used to model the degradation. It is assumed that the parameters are affected by epistemic uncertainty, which is caused by incomplete knowledge about a model or its parameters. Besides, when there are no enough data to estimate the unknown parameters, the probability theory cannot be applied, so the alternative evidence theory, which is also known as Dempster–Shafer theory, may be used to describe epistemic uncertainty. Here, the judgments and mental estimates of some experts are gathered in interval form to get the belief measures for a reliable system. Then, the unknown parameter is estimated. A numerical algorithm is proposed to obtain a high-quality maintenance policy. Finally, a numerical example is presented to illustrate the procedure.

Keywords